Help With Tube Mic Pre

GroupDIY Audio Forum

Help Support GroupDIY Audio Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.

cpsmusic

Well-known member
Joined
Dec 3, 2013
Messages
292
Location
Melbourne, Australia
Hi Folks,

About twenty years ago my father built me a tube mic preamp. He was mostly a radio and HiFi guy and so when I started asking for guitar amps and then later, studio gear, it was a bit outside of his area of knowledge. One of the things I got him to build for me was a stereo tube mic pre. I wanted something that I could overdrive as required but other than that I left the design up to him. When the unit was built it had a pretty serious hum problem. Anyway, the unit was basically put in the "to be looked at" basket and eventually found its way into storage where it's been up until about a year ago. Sadly, my father passed away in 2016 so the hum problem was never fixed.

Fast forward to about a year ago when I pulled the unit out of storage. I had a tech look at it however because the unit was point-to-point it was going to be too expensive to get repaired (hourly rates!). I do remember the tech opening up the pre and then looking at me saying "you do realise that most things in there are more than forty years old"! My father was someone who never threw anything away and would often re-use and cannibalise stuff so some of the components could easily have been from the 1950s!

I thought I'd have a go at fixing the unit myself. Just FYI I'd say I have an intermediate level knowledge of electronics. The last project I did was a rebuild of a tube guitar amp.

So to the mic pre and its problems. At the moment the level in one channel is very low while the other channel has a noticeable hum. When the tech looked at the unit he said he suspected the problem was an "open plate" although I'm not sure what this means in terms of troubleshooting.

I've attached a schematic of the unit as well as a picture. As can be seen, the point-to-point wiring around the tubes makes it hard to see what's going on.

Any suggestions as to where to start, what to look for, or try are welcome.

One thing I have tried is replacing the tubes (I had three new 12AX7s that were from a working guitar amp) and it didn't change anything. The hum problem was still there.

 

Attachments

  • IMG_2660.jpg
    IMG_2660.jpg
    762 KB · Views: 68
First things I notice are that 12AX7 and 12AU7 are very different tubes and you cannot just use either in the same circuit. The pencilled tube types are 12AU7 in all three stages but someone has later added 12AX7 in ink. Was that you or your Dad? At first glance it looks like the tubes are meant to be 12AU7. This will give about 20dB gain per stage, plus probably another 20dB from a 1:mad:10 input transformer for a total of 60dB which seem to me to be in the right region.

The second thing I notice is the final stage, the cathode follower seems to be incorrect; it is not biased properly. I suspect the schematic is not accurate here.

Everything else looks basically OK  to me. However, you should make sure the chassis is connected to mains earth for the sake of safety.

Cheers

Ian
 
Looks decent enough,

But:

..on your input XLR's, pin2 is grounded, pin1 and 3 are signal. This is very wrong.

To get rid of hum, I would start out by determining where it enters the circuit. Put in only the last tube, then the next-last, then the first. You have metal top and bottom lids for the chassis/box? That is needed. And make sure your 0V has actual connection to your chassis.

Short the input after the input transformer(s)- if hum goes away with shorted input, you have pickup in your input transformer. Try moving the power transformers physically away and see if that changes anything significantly

second, I'd put a hum trimmer in heater circuit ground (center tap-to-gnd)

Third, if everything else fails, I'd stabilize (regulate) the HT supply - see e.g. G9 for a simple way to do this

Last, I don't like the way of it's phantom at all

Deriving phantom power from +300V by resistor divider is a catastrophe waiting to happen. You don't want it this way. Again, look at G9 for a simple way to get this type of voltages.

Also, phantom feeding from center of transformer primary is - although not technically wrong - not a good way to do it. And schematic does not show dropping resistors (!) You shouldn't connect two phantom powers in parallel like that without individual  current limiting resistors. Use two 6K8's on each input like everyone else.

I don't think you will want the direct connection between center taps on input transformers in any scenario, better make that two individual phantom switches.

Also, I don't think you'll want to have VU-meters hanging permanently on the cathode-follower output - this is a significant but unnecessary source of (often-unwanted-type) distortion. In stead use a couple of opamps to buffer for metering.

(not gone all the way through this!)

/Jakob E.
 
ruffrecords said:
First things I notice are that 12AX7 and 12AU7 are very different tubes and you cannot just use either in the same circuit. The pencilled tube types are 12AU7 in all three stages but someone has later added 12AX7 in ink. Was that you or your Dad? At first glance it looks like the tubes are meant to be 12AU7. This will give about 20dB gain per stage, plus probably another 20dB from a 1:mad:10 input transformer for a total of 60dB which seem to me to be in the right region.

The second thing I notice is the final stage, the cathode follower seems to be incorrect; it is not biased properly. I suspect the schematic is not accurate here.

Everything else looks basically OK  to me. However, you should make sure the chassis is connected to mains earth for the sake of safety.

Cheers

Ian

That's my Dad's writing. I seem to recall that at the time the 12AU7s weren't distorting enough (did I say that played guitar in a hard rock band and that this was my first mic pre  :D). Think I might go back to using the 12AU7s at least until I can get the unit working.

Chassis ground was the first thing I checked and it's OK.

Also, I built myself a dim bulb current limiter which I've used during the initial tests.

Cheers,

Chris
 
gyraf said:
Looks decent enough,

But:

..on your input XLR's, pin2 is grounded, pin1 and 3 are signal. This is very wrong.

To get rid of hum, I would start out by determining where it enters the circuit. Put in only the last tube, then the next-last, then the first. You have metal top and bottom lids for the chassis/box? That is needed. And make sure your 0V has actual connection to your chassis.

Short the input after the input transformer(s)- if hum goes away with shorted input, you have pickup in your input transformer. Try moving the power transformers physically away and see if that changes anything significantly

second, I'd put a hum trimmer in heater circuit ground (center tap-to-gnd)

Third, if everything else fails, I'd stabilize (regulate) the HT supply - see e.g. G9 for a simple way to do this

Last, I don't like the way of it's phantom at all

Deriving phantom power from +300V by resistor divider is a catastrophe waiting to happen. You don't want it this way. Again, look at G9 for a simple way to get this type of voltages.

Also, phantom feeding from center of transformer primary is - although not technically wrong - not a good way to do it. And schematic does not show dropping resistors (!) You shouldn't connect two phantom powers in parallel like that without individual  current limiting resistors. Use two 6K8's on each input like everyone else.

I don't think you will want the direct connection between center taps on input transformers in any scenario, better make that two individual phantom switches.

Also, I don't think you'll want to have VU-meters hanging permanently on the cathode-follower output - this is a significant but unnecessary source of (often-unwanted-type) distortion. In stead use a couple of opamps to buffer for metering.

(not gone all the way through this!)

/Jakob E.

Thanks, that's plenty for me to go on with. Yes, there's a top and bottom for the case. What do you mean by "moving the power transformers physically away" - do you mean remove the power supply from the case?

Cheers,

Chris
 
I see some effort has been made to screen off the mains transformer , generally though for best hum performance an off board PSU is the way to go , 
Orientation of input transformers relative to mains transformer can make a huge difference also , you may gain some improvement by careful positioning , rotating for minimum hum and screening  , of course all of that comes after you've made sure all the active portions of the circuit are running properly .

An in/out (of circuit) switching arrangement for the metering would be an easy thing to add . I always set levels by ear on my tube gear allowing in some extra distortion if thats whats needed , a moving coil meter doesnt give much of an indication of the subjective effects of distortion in any case so the ear is the best way of dialing in whats tollerable or wanted . Source material has a huge bearing on  harmonic content generated and what the ear percieves as acceptable wont be at the same %THD right across the frequency spectrum. Many of us here use REW to view the harmonic spectrum produced in equipments , it will give an individual % level for each harmonic generated ie 2nd 3rd 4th etc which is usefull .

A cold check with a multimeter of the resistor values might be a good place to start , coupling caps might also be suspect,
any sign of positive volts at the grid after the coupling cap are a likely sign of trouble  ,especially with older paper/foil types. Modern electrolytic caps like the 47u and 100uf in the psu are much smaller these days ,so its usually worth putting in larger values for lower hum on the HT rail  as well as keeping wiring more tidy.

Ive always found replacing components on hard wired stuff easy , I dont see the need to desolder ,just clip out the suspect component and replace it . The fault finding and or mods to improve things will take time to work out ,so thats really where the time and expense come in.

Your first stage has an anode load resistance of 100kohms then the second stage has 250kohms , your second stage will inevitably have less output voltage swing available than the first, but maybe that was part of the plan to create more distortion.
Best of luck with your project.





 
gyraf said:
Last, I don't like the way of it's phantom at all

Deriving phantom power from +300V by resistor divider is a catastrophe waiting to happen. You don't want it this way. Again, look at G9 for a simple way to get this type of voltages.

Also, phantom feeding from center of transformer primary is - although not technically wrong - not a good way to do it. And schematic does not show dropping resistors (!) You shouldn't connect two phantom powers in parallel like that without individual  current limiting resistors. Use two 6K8's on each input like everyone else.

I don't think you will want the direct connection between center taps on input transformers in any scenario, better make that two individual phantom switches.

/Jakob E.

Would it be possible to get the phantom power from the 15VAC that's being used for the relays?
 

Latest posts

Back
Top